MEASURING THE TOTAL COST OF INJURY IN NEW ZEALAND:

Transcription

1 Business and Economic Research Limited Report to: MEASURING THE TOTAL COST OF INJURY IN NEW ZEALAND: A REVIEW OF ALTERNATIVE COST METHODOLOGIES BERL #4172 Prepared by Mark Goodchild Kel Sanderson Ganesh Nana All work is done, and services rendered at the request of, and for the purposes of the client only. Neither BERL nor any of its employees accepts any responsibility on any grounds whatsoever, including negligence, to any other person. While every effort is made by BERL to ensure that the information, opinions and forecasts provided to the client are accurate and reliable, BERL shall not be liable for any adverse consequences of the client's decisions made in reliance of any report provided by BERL, nor shall BERL be held to have given or implied any warranty as to whether any report provided by BERL will assist in the performance of the client's functions. Business and Economic Research Ltd Level 5, 108 The Terrace PO Box Wellington New Zealand phone: fax: web-site: innovate@berl.co.nz

2 MEASURING THE TOTAL COST OF INJURY IN NEW ZEALAND: A REVIEW OF ALTERNATIVE COST METHODOLOGIES 1 EXECUTIVE SUMMARY WHAT IS THE TOTAL COST OF INJURY DIRECT COSTS INDIRECT COSTS INTANGIBLE COSTS WHAT ARE THE CORE METHODS OF VALUATION DIRECT COST METHODS INDIRECT COST METHODS INTANGIBLE COST METHODS TOTAL COST METHODS A SUMMARY OF THE CORE METHODS HOW HAVE THE CORE METHODS BEEN HYBRIDISED COMBINING EARNINGS WITH WTP INCLUDING HEALTH STATUS INDICES MONETISING HEALTH STATUS INDICES WHAT ARE THE POTENTIAL WAYS FORWARD DIRECT COSTS INDIRECT COSTS INTANGIBLE COSTS A POTENTIAL FRAMEWORK REFERENCES Measuring the Total Cost of Injury in New Zealand

3 1 EXECUTIVE SUMMARY This report has been prepared by Business and Economic Research Ltd (BERL) to assist the Department of Labour to develop a robust and practical methodology for measuring the cost of injury in New Zealand. The purpose of this report is to review the methods of valuation that have been applied elsewhere, and to indicate how these methods might be applied in New Zealand. Thus, this report provides a basic framework for measuring the total cost of injury in New Zealand. Note that this report focuses on methods used to value the cost of injuries as opposed to methods for assessing whether programs or new investments designed to mitigate these costs yield net benefits from an economic or societal perspective. However, these methods of assessment (Cost-Benefit, Cost-Utility, etc) all require some measure of loss in order to assess the merits of such programs. The framework outlined in this report therefore recognises that different organisations can have different policy objectives and methods of assessment. 2 Measuring the Total Cost of Injury in New Zealand

4 2 WHAT IS THE TOTAL COST OF INJURY This section describes the total cost of injury with cost elements categorised into direct, indirect and intangible costs, which is consistent with the literature in health economics. These costs are not limited to the individual and include costs that filter through to other aspects of economic and societal wellbeing. The below table shows the composition of total cost across two dimensions with the direct, indirect and intangible cost components identified according to their potential impact on economic and societal wellbeing. Thus, total cost can be defined as the sum of all direct, indirect and intangible costs across this socioeconomic dimension. Table 2.1 The composition of the total injury costs TOTAL INJURY COSTS Individual Family Employer Economy Societal Direct Costs Accident costs (i) Medical costs Non-medical costs Indirect Costs Absenteeism Productivity losses: i) reduced activity/ability ii) reduced participation Family worker substitution Worker replacement/substitution Taxation Intangible Costs Loss of life Loss of life expectancy Loss of quality of life Physical suffering Mental suffering (i) not strictly a cost of injury Source: BERL Total cost covers a range of social and economic interactions which in turn depend upon a number of variables including the level of functional impairment caused to the injured person. This can be regarded as the temporary and permanent change in the health status of an individual following an injury event. Change in the health status of the individual plays a central role in determining the impact of an injury event, with the magnitude of the change in individual health status affecting the ability of the individual to work and the permanence of intangible costs such as pain and suffering. 3 Measuring the Total Cost of Injury in New Zealand

5 The direct, indirect and intangible costs of injury events are discussed in more detail in the rest of this section. Some general observations are also made regarding the impact of these injury costs. However, these impacts are not discussed in more detail until section 4 of this report. This is because the methods of valuation (in section 3) are based upon different theoretical frameworks, which lead to different and sometimes opposing views regarding the impact of injuries, and therefore the total cost of injury events. 2.1 DIRECT COSTS Direct costs are expenditures and damages relating to the occurrence and the prevention of an injury. These can be categorised as accident costs, medical costs, and non-medical costs as shown in the table 2.1 above Accident Costs Accident costs are comprised of physical property damages as the result of an incident and include assets such as vehicles, buildings and equipment. Although these damages are strictly accident costs as opposed to injury costs, these costs are often included in Economic Impact Assessments (EIAs) of events or accidents. The definition of an accident cost can be widened to include the value of other physical assets such as the environment. Thus, events such as environmental disasters could be assessed through estimating the costs from both the damage to physical assets as well as the cost of human injuries and fatalities. The impact of accident costs can therefore span across the socioeconomic dimension from the individual too societal wellbeing (eg the environment). The importance of accident costs will tend to differ according to the event that is being assessed. In some cases such as natural disasters, accident costs will be quite significant especially when the resulting losses also lead to downstream impacts upon production. Downstream impacts relate to the loss of economic production (ie GDP) that occurs as a result of these assets being unavailable for use in the production process. An illustration of how accident and injury costs can be incorporated is given in the box below, which assesses the economic impact of industrial fires in New Zealand. 4 Measuring the Total Cost of Injury in New Zealand

6 Profile of Industrial Fire Costs (BERL, 2002) The report entitled An Economic Assessment of Industrial Fires which was completed for the New Zealand Fire Service Commission (NZFSC) provides an example of including injury costs in economic impact assessments. In this report, building damages from industrial fires was combined with estimates of NZFSC operational costs, downstream production losses, and injury costs. The downstream production losses were estimated from company insurance claims data for business interruption in combination with input-output tables, which together gave a measure of the resulting losses in industrial production (ie GDP). Injury costs were estimated using injury data contained in the Fire Incident Reporting System (a database maintained by Fire Service personnel) and the Land Transport Authority s measure of the Value of Statistical Life. Note that the LTSA s measure of the Value of Statistical Life is discussed in section 3 of this report. The composition of economic costs from industrial fires for the year 2000 is shown below. Material damages, which includes insured assets such as plant and buildings, stock, machinery and equipment accounts for about 40% of the total economic costs and can be thought of as direct accident costs. The operational costs of the NZFSC were also high at about 30% of the total economic costs. However, the cost of fire related injuries were quite low in this study, reflecting the fact that relatively few industrial injuries occurred. Cost Component $Mn Material damages 36.1 NZFSC costs 23.3 Production losses 18.4 Injuries and fatalities 8.5 Total costs 86.3 Source: An Economic Assessment of Industrial Fires in New Zealand. BERL While direct accident costs can be quite significant, the purpose of this report is to focus on the methods of valuing injury. Therefore, accident costs have been put to one side for the rest of this report, except to note that the valuation methods applied in other fields such as environmental economics have often been adopted for valuing injuries in health economics. These methods are discussed in section 3 of this report Medical Costs Direct medical costs are comprised of expenditures on goods and services relating to the medical care of patients, and capital investment in facilities, staff training and research. Medical goods and services are those used in the prevention, detection, transportation, treatment, and rehabilitation of the patient. 5 Measuring the Total Cost of Injury in New Zealand

7 While the case for medical goods and services to be included as a direct cost is obvious, the case of including capital investment is perhaps less apparent. However, the inclusion of capital investment is based on the concept of opportunity cost - where investment in these health services constitute a missed opportunity to use these expenditures for other investment or consumption activities. Hence, overhead medical charges such as interest expenditure and administration costs are also included as direct medical costs. There is a range of issues regarding the correct process for estimating direct costs, with the biggest problem being that prices charged in health care to consumers are distorted and rarely reflect true economic value. The interaction between insurance plans and the regulatory system yield a gap between accounting and economic (opportunity) costs. To calculate total health care costs in New Zealand s multiple payer system of health care financing, contributions from several public and private sources need to be considered and corrected for double accounting. In New Zealand, few medical services are provided on a strictly commercial basis. The ACC does, however, have a set of standard or maximum charges where costs are quite explicit and so these conceptual issues can be addressed. Specifically, it may be possible to obtain adequate estimates of average direct costs over a range of relevant medical procedures. The ACC also has a rich database of injuries and treatment costs, which can assist in the estimation of direct costs. The main elements missing from these costings are those medical and non-medical expenditures by individuals and employers that are not covered by the ACC. Some of these expenditures may be available from the private medical insurers through their policies, some of which provide full cost reimbursement. The ACC scheme can be regarded as a 24-hour no-fault compensation scheme and in this sense the medical costs of injuries have been socialised in New Zealand. Hence, the impact of actual medical costs is not borne by the injured person but by individuals and employers in aggregate through the ACC levy system. Because this system is socialised, the impact of direct costs is more complicated. But generally speaking, the impact of the injury costs via the ACC levy will depend on the ability of enterprises to pass-on the ACC levy to employees or to consumers. However, the issue of who bears the cost of ACC does not affect the level of actual total costs involved. 6 Measuring the Total Cost of Injury in New Zealand

8 2.1.3 Non-Medical Costs Direct non-medical costs consist of expenditures on goods and services incurred as a result of the injury event, but which are not related to the medical care of the patient. Non-medical costs include those of non-emergency transport, informal care, household help, vocational and family counselling, and alterations to the home and business. Table 2.2 presents a summary of the direct medical and non-medical costs considered in five studies. This table has been adapted from a report by Beeck & Mulder (1998). The authors note that large differences exist in the selection of direct cost elements and that in some studies important costs are lacking. They also note that these differences could reflect international differences in the structure and financing of medical care. It is also evident from these studies that the data required will be dependent on the nature of the injury and the extent to which on-going care or rehabilitation services are required. Table 2.2 direct medical and non-medical cost elements by various studies DIRECT MEDICAL AND NON-MEDICAL COSTS Den Toom & Rice et al COST SWOV CPSC Schuurman (1998) (1998) (1998) (1996) (1990) Emergency Services First aid Transport Hospital Care Tests Treatment Medical personnel Facilities Rehabilitation Outpatient treatment Re-hospitalisation Non-hospital Care Physician care Physical therapy Nursing home Attendant care Dental care Medication Aids & appliances Non-medical Costs Home adaptations Special transportation Vocational rehabilitation Educational rehabilitation Administration insurance costs Source: BERL, adapted from Beeck, E & Mulder, S. (1998) 7 Measuring the Total Cost of Injury in New Zealand

9 2.2 INDIRECT COSTS Indirect costs can be defined as the output losses resulting from an injury event. For the individual these costs consist of output losses due to morbidity or mortality. In the case of morbidity, output losses can be the consequence of reduced productivity and reduced participation in the workforce. In addition, injuries may cause unwanted job changes for the individual and lost opportunities for promotion and education. Generally speaking, the size of these output losses will tend to relate to the individual s level of functional impairment over the short, medium, and long term. The figure below shows how functional impairment can impact upon an individual s participation in the workforce, their ability to earn and therefore, the broad magnitude of the indirect impact upon the individual. Figure 2.1 Disability pathways and the labour market However, there remains considerable debate within the economics profession regarding the impact of these indirect costs. While this issue is discussed in more depth in section 3 of this report, the broad question is whether or not these losses for the individual have an impact upon the output of the firm and hence to total economic output. This question reflects the different theoretical approaches taken in the Human Capital Method (HCM) and the Friction Cost Method (FCM). Nonetheless, the figure above does illustrate the relationship between functional impairment and its influence on the individual s return to the workforce, and is especially relevant for interventions relating to rehabilitation, occupational counselling, etc. 8 Measuring the Total Cost of Injury in New Zealand

10 Indirect costs include the output losses of other family members in cases where informal patient care replaces non-labour market activities such as housekeeping which otherwise would have been performed. In the case of long-term injury, family members may need to enter the workforce or work longer hours to compensate for lost household earnings, a situation described as family worker substitution in table 2.1. Although it is possible to value non-labour market output losses and family worker substitution using standard methods, these costs are seldom included in empirical studies due to the lack of accurate information. There can also be indirect costs to the employer as caused by absenteeism resulting from an injury. As an illustration, the opposite effect is discussed in below. Absenteeism Benefits from Improved Health & Fitness BERL s study on the Economic Impacts of Sport for the Hong Kong Sports Development Board (HKSDB) included a review of the indirect benefits of improved health and fitness upon productivity and absenteeism. The table below shows the impact on absenteeism as recorded by a number of studies on the introduction of health and fitness programs at the firm or organisational level. Most of these studies show quite a strong correlation between the introduction of a program and lower absenteeism, with the link presumably being both physical and psychological for program participants. A number of these studies also reported positive impacts on employee productivity levels from the introduction of health and fitness programs. Again, this correlation was linked to physical and psychological factors such as improved cognitive functioning, reaction time, stamina and attitude. Source: The Economic Benefits of Sport. BERL Measuring the Total Cost of Injury in New Zealand

11 2.3 INTANGIBLE COSTS Intangible costs are the reductions in the welfare of people due to physical, mental and behavioural problems as the result of an injury event. There are a number of adverse health outcomes typically associated with reductions in welfare including, mental and physical suffering, the quality of life, and the length of life expectancy. Because these reductions in health status also affect individual welfare, economic theory suggests that people are willing to pay to restore their health or to avoid the risk of such outcomes. Injuries can also impact upon the welfare of family members through deteriorations in mutually dependent relationships with the injured person. These intangible costs include the mental suffering of siblings and spouses, and social problems such as divorce. With mental illnesses or drug addition, these social problems can include other costs such as bankruptcy, violence and criminal convictions. The intangible costs of injury are primarily borne by the individual or family members. However, the concept of consumer willingness to pay is also interesting when assessing the impact of intangible costs because it considers how people value the risk of avoiding such costs. This approach suggests that people s sense of welfare is also a function of altruistic concerns for the welfare of others in society (eg societal wellbeing). A recent study by Strand (2001) provides some interesting insights into intangible costs by estimating the composition of peoples Willingness To Pay (WTP) to avoid certain health risks. His results showed that about one-third of the total WTP measured in his research was motivated by pure self-interest, about one-half was attributable to concerns for the family and the remaining share was attributable to altruistic concerns for other people in society. Other studies have found varying degrees of concern for individual wellbeing, family wellbeing, and societal wellbeing. 10 Measuring the Total Cost of Injury in New Zealand

12 3 WHAT ARE THE CORE METHODS OF VALUATION 3.1 DIRECT COST METHODS Direct costs can be measured by using either a top down or bottom up approach to allocating health expenditures across groups. The incidence (or bottom up) method has the advantage of enabling costs comparisons at a fine level of aggregation compared to the prevalence (or top down) method, but has the disadvantage of requiring detailed data sources which can be time consuming to generate The Prevalence Method In the prevalence method, direct health care costs are assigned top down to each injury classification. Direct costs are those total expenditures considered to be relevant within a certain health sector, which is then distributed to sub-categories of patients according to their respective share in total prevalence. Usually, total prevalence is measured using a consistent indicator such as the share of total hospitalisation days. Note that the prevalence method provides an estimate of average as opposed to marginal costs. Thus, it is not possible to estimate the true costs incurred from treating a single or additional patient. In other words, some caution is warranted when interpreting the level of health care savings that may result from programs designed to reduce the occurrence of specific injury groups. This is especially be the case if the indicator of prevalence is an inexact measure of marginal costs because for example, the resource required even in similar groupings of injury patients differs by a significant amount The Incidence Method The incidence method takes a bottom up approach to estimating costs and is based on the resource costs of treating well-defined patient groups at a fine level, which are then aggregated towards the desired level of patient grouping. The incidence method has the advantage of enabling researchers to compare patient costs at a fine level of aggregation when compared to the prevalence method, but has the disadvantage of requiring detailed data sources, which are generally not available and can be time consuming to generate. Generally speaking, the incidence method may be less likely to gloss over distortions caused through the aggregation process, especially when the gaps between marginal and average costs between patient groups are quite large. 11 Measuring the Total Cost of Injury in New Zealand

13 The Operational Costs of the NZFSC using the Prevalence Approach In BERL s report entitled An Economic Impact of Industrial Fires the prevalence method was used to allocate the operational costs of the New Zealand Fire Service Commission to industrial fire incidents. Firstly, the relevant level of expenditure had to be identified. The Fire Service s annual report showed that the cost of emergency response was just 9% of the total $172 million expenditure on operational costs, and that much of this total was actually related to operational readiness. However, these emergency responses could not have been made if the Fire Service is not kept in a state of operational readiness and thus, the total operational costs (eg $172 million) were taken as the relevant figure. A fairly simple prevalence approach would have been to allocate these costs across all incidents on the basis of a standard indicator. For example, the share of each type of incident relative to all emergency incidents. With industrial fires accounting for about 8% of all active incidents this approach would have allocated about $13 million in operational costs to industrial fires in New Zealand. However, would this indicator provide a fair representation of the quantum of resource used across different types of emergency incidents? Instead of using the simplified approach above, this study allocated operational costs according to the broad quantum of resource used in response to various emergency incidents. The table below shows a summary of how this allocation was made. Most importantly, the table shows that the average industrial fire required 81 appliance minutes compared to 47 appliance minutes for all active incidents. Thus, on average, the quantum of resource delivered to industrial fires was significantly higher than all active emergency incidents. With industrial fires accounting for about 14% of the total resource delivered (eg total appliance hours), the study estimated that about $23 million in operational costs should be allocated to industrial fire indicants. Resource Delivered Number of Incidents Average Appliance Minutes Resource Delivered (appliance hrs) % of Resource Delivered Allocation of Operational Costs ($000) False alarms and good intent calls 24, ,721 Non-fire incidents 14, ,989 37% 63,254 All fire incidents: 20, ,375 64% 110,024 Other / miscellaneous fires 10, ,330 27% 46,417 Residential fires 7, ,372 23% 40,353 Industrial fires 2, ,672 14% 23,253 All Active incidents 34, , % 171,974 Source: BERL (2002) This approach also highlighted a number of interesting questions regarding the operational costs of the NZFSC and the potential impact of reducing (or targeting reductions in) certain types of emergency incidents. For example, how might a long-term reduction in industrial fire incidents, which are resource intensive, impact upon the level of operational readiness (eg capacity) that is required by the NZFSC? And how would this change affect the level of the NZFSCs operational costs in total and on average across all emergency incidents? Source: An Economic Assessment of Industrial Fires in New Zealand. BERL Measuring the Total Cost of Injury in New Zealand

14 3.2 INDIRECT COST METHODS Methods of valuing only indirect costs focus on measuring the output losses caused by an injury event through examining the relationship between an injured person and the workforce, and thus the impact upon the aggregate economic output The Human Capital Method The Human Capital Method (HCM) is the traditional valuation method used to measure the indirect cost of injury events. Using the HCM approach, indirect costs are equivalent to the production that would have been produced in the absence of the event, including the loss of productive life years due to fatal injury events. These production losses are measured as the discounted stream of future income foregone by the individual. This future income is discounted because the future value of capital is less than the current value, and because of time preference. The fundamental presumption of HCM is that people are a valuable economic resource. People represent a source of productive capital that is used to produce a future stream of production. An injury reduces or destroys that future stream of production and therefore these events reduce the total quantum of human capital available to the economy and the economy s capacity to produce goods and services. At this level, the HCM assumes that this loss of capacity reduces both actual and potential future production. Hence, the total indirect cost of injury at the macroeconomic (economy-wide) level is simply equivalent to the aggregation of discounted future income losses from all injury events. Given these assumptions, the impact of an injury event on individual earnings is central to the HCM approach. At this level, the HCM approach measures the discounted future value of income foregone from temporary and permanent reductions in productivity as a result of the injury. Therefore, the HCM rests on the assumption that earnings reflect productivity. However, this is not to say that each employee receives the value of his or her personal contribution to output, but that each worker receives the value of output added by the marginal (or last hired) worker. The HCM is frequently criticised because some groups of individuals are undervalued relative to other groups due to the observation that human capital values are higher for men than women, for whites compared to minorities, and for the middle aged compared to the young and elderly. The premise underlying this criticism is that earnings for these groups do not necessarily reflect the relative value of their marginal product. 13 Measuring the Total Cost of Injury in New Zealand

15 The HCM approach has a long history and several derivatives have been developed over time. One derivative is to measure an individual s net input into the aggregate economy by measuring earnings net of consumption. This is based on the concept that net input is equivalent to the individual s contribution to total savings, which equals the individual s contribution to total investment in the economy. But generally speaking, consumption is no longer netted out of HCM estimates. Most estimates have included earnings gross of taxes (to reflect the social loss from foregone earnings), and have tended to exclude non-labour market income with the rationale for this exclusion being that non-labour income would not be lost to society even with the death of the individual. A second derivative of the HCM has been developed in response to alternative methods of estimation such as the Contingent Valuation approach. These methods have tended to focus on individual welfare as opposed to aggregate economy-wide costs. In response, a number of studies have attempted to provide a more individualised estimate of human capital. The leading authors here are Landefeld and Sekin (1982) who individualised their HCM measure by estimating earnings net of taxation, including non-labour market income, using an individual (rather than social or market) discount rate and including a risk aversion factor. However, a significant criticism of the HCM approach (including the derivatives above) is that intangible costs remain outside of the scope of study. Given that intangible costs can be a significant component of individual welfare, especially with severe long-term injuries, this omission can have a significant impact on the magnitude of total costs. Nonetheless, the justification of the HCM is not that it measures the full value of life but that it provides a useful measure of indirect costs. This is an important component of total cost and is of considerable interest to policy makers and so forth Whole Economy Cost of Human Capital The HCM presumes that people are a valuable economic resource, or unit of productive capital that produces a future stream of production. Injury reduces the future productive capacity of that resource and thus the present value of that productive human capital. The economic implication of this presumption is that the reduction in production from the injured person cannot readily be replaced by some presently unemployed human resource being mobilised to maintain the past level of production. The contra situation (that unemployed labour can in all cases be mobilised) is assumed in the Friction Cost Method (FCM) discussed below. 14 Measuring the Total Cost of Injury in New Zealand

16 In situations where the HCM presumption is an accurate representation, the capital (eg capacity, skills, etc) of the injured person is not readily replaceable from unemployed resource, and thus the future production from that person in their previous occupation is reduced or lost. When this occurs there will be upstream and downstream multiplier effects, representing reduced demand for inputs used in that work position and reduced income available for consumption spending within the economy. Thus, the impact upon economy-wide production will be more than the lost production of the injured person. The total quantum of the reduced economy-wide production can be estimated in a partial equilibrium sense using multiplier analysis. When the indirect cost of injury is being measured for a large number of individuals (in a certain industry, or in certain skill sets or across the economy) the aggregate impact of these injuries would be a significant reduction in the total labour supply. Because this reduction in labour supply would lead to an increase in the price of labour, the indirect costs of injuries should then be measured in general equilibrium terms, by assessing the impact on the economy of changes in the supply and price of labour. This indirect cost can be measured by running a General Equilibrium Model (GEM) with a base case scenario and then a scenario with the labour supply reduced due to injury. In this model the injury cost to the economy would be the difference in GDP achieved between the two scenario runs of the model. BERL has a GEM of New Zealand with 49 industries and 40 occupations in the core matrix. This model enables simulations of the supply of labour in various industries and occupations to be tested as to the re-allocation of labour amongst all industries and thus, the level of GDP which can be achieved with the reduced labour available. This method would then give an estimate of the economy-wide indirect cost of injuries. 15 Measuring the Total Cost of Injury in New Zealand

17 A Measurement of Indirect Health Benefits Measurement of indirect costs of injury requires an assessment of the loss of production as a result of injury at the individual and/or the economy-wide level. There are a number of studies providing evidence of the positive impact on the economy from a gain in production due to improved health and improved work functionality. These studies (some of which are listed in the table below) have shown increases in productivity associated with introduction of corporate exercise programs. A further example comes from a Canadian Government discussion paper that estimated that a 25% increase in the 1995 physical activity participation rate increased labour productivity in the whole economy by between 0.25 and 1.5%. And finally, a 1992 Canadian National Workplace Survey showed that more than 60% of companies with fitness programs realised improved productivity, reduced absenteeism, reduced employee turnover and led to fewer accidents. Working from this base of information, a study by Tasman Asia Pacific, Ernst & Young (1998) simulated the effect on the Australian economy of increased worker participation in sport and recreation to participation of 100%. They assumed this would result in a 4% sustainable increase in labour productivity. The simulation was made using a general equilibrium model (GEM) to estimate the gain to the economy. The purpose of the simulation was to highlight the impact and implications of productivity improvements on the Australian economy as a whole. A 4% increase in labour productivity equated to a 2.8% overall increase in Australia s productivity, which was distributed across all industries in the model, in order to estimate the change in industry output and other macroeconomic consequences. Resulting GDP increased by 3.3%, consumption by 4.1%, investment by 3.7%, exports by 4% and imports by 4.2%. The top 25 industries would increase activity by 5% to 11%. Employment would increase by 1.4%. Relevance to New Zealand In New Zealand, if injury effects upon productivity (for specific or total injuries) were available for specific occupations or industries, the BERL GEM could be used to estimate the economic costs of injury in the same way as Ernst and Young used a GEM above to estimate the economic benefits of sport. In the injury case it would be possible to estimate the indirect costs of injury at the economy level and thus provide supporting information on the indirect impacts of health or injury programs. Source: The Economic Benefits of Sport: A Review. BERL Measuring the Total Cost of Injury in New Zealand

18 3.2.3 The Friction Cost Method The Frictional Cost Method (FCM) can be regarded as a critique of the Human Capital Method (HCM) of measuring indirect costs. The proponents of the FCM argue that the HCM overstates indirect costs at the economy-wide level, because the HCM measures potential as opposed to actual production losses. The FCM argues that the actual production loss is limited because several mechanisms in the labour market move to limit the impact of individuals on actual production. Over the short-term, the work that would have been undertaken by the injured individual can be postponed or reallocated to colleagues. Over the long-term, unemployed workers can substitute for those injured workers. Hence, the FCM measures indirect costs as only the temporary production loss due to an injury event and the financial expenditures required in recruiting and training new employees. These costs could still be quite considerable. For example, if a top executive or professional is lost to the economy, an unemployed person may be trained to move onto the bottom rung of the ladder and a chain of people each take a step up the ladder to fill the vacant professional post. The frictional training cost at each step could therefore be quite considerable. The economic and labour force scenarios represented by HCM and FCM estimations of injury cost are quite different and therefore they measure completely different aspects of labour market operation. The HCM assumes that the capacity lost through injury cannot be replaced and therefore measures the value of future production lost. The FCM scenario is that the injured capacity lost can be quite quickly replaced by recruiting and training another (presently unemployed) person to the position left vacant by injury. The FCM therefore concentrates on measuring temporary production loss only, costs of recruitment and training (formal and also informal by colleagues on-the-job) to give the indirect cost of injury. In this scenario, the assumed infinite availability of unemployed labour enables future production to be maintained and therefore there are no future costs to the whole of the economy, and no need for multiplier or GEM analysis. The HCM and FCM approaches make assumptions at two extremes in relation to the capacity of the economy. The HCM measure assumes that the value of labour lost is not replaced and thus implicitly that the resources within the economy are operating at full capacity with no possibility to replace human capital losses. Alternatively, the FCM method implicitly assumes that there is an endless pool of unemployed people who can be trained to substitute for anyone lost to the production process. 17 Measuring the Total Cost of Injury in New Zealand

19 The business cycle is also relevant. In an economy growing strongly, the rate of growth is likely to become constrained by total skill availability and the HCM will be closer to the production loss than the FCM. The skills and human capital that has been created in the economy will be fully utilised and the loss of some of this human capital will reduce actual total capacity. In a sluggish economy, with substantial unemployed labour, the FCM is likely to be a closer representation of indirect costs, because labour capacity is unlikely to be a factor limiting long-term productive capacity. Empirical studies using the FCM naturally produce estimates of indirect costs that are much lower than comparable HCM estimates. Koopmanschap et al (1995) estimated the indirect costs of disease using the FCM to be 1.2% of net national income compared to 18% when measured by HCM. A study by Van Beeck (1997) produced FCM estimates, which were about 20% of the HCM. An Australian Measure of Job Change Frictional Cost The Australian Productivity Commission (APC) measured the cost of job-change as a result of problem gambling - namely the Frictional Cost. The three elements they identified were: Loss of income over the period of unemployment before a new job is found; The financial cost of the job search; and The cost to the employer of finding and training a replacement. The APC noted the effect of benefits in transferring a substantial part of the loss of income from individuals to the taxpayer, or the economy or society. Their estimates for the three elements were as follows: Period between jobs of 6 weeks, average earnings of $743 per week gives loss in income of $4,300. Cost of job search was taken to be $2,357, which an earlier paper had quoted as being approximately half of the cost reported by major job search firms. Cost of staff replacement by the employer. This conceptually included hours of staff spent hiring, hours of formal training, hours of informal training by management and hours of informal training by co-workers. Using other sources, the APC assumed employer search and replacement cost equals 10% of annual salary, or $3,862. Additionally the APC noted that some of the people who change jobs also receive unemployment benefits transferred from government. This amounted to an average of $1,482 per person overall. The APC estimate of frictional cost of job change is thus a total of $12,000 on average for all people who change job. Source: Australia s Gambling Industries. Australian Productivity Commission Measuring the Total Cost of Injury in New Zealand

20 3.3 INTANGIBLE COST METHODS Methods of valuing only the intangible costs of injury, namely the costs of pain and suffering focus upon analysing the choices made by people as individual or collective decision-makers. The quantum of these intangible costs can be determined in monetary terms by analysing the court awards made for injury compensation or can be taken from compensation schedules for injuries established by public administrations. Alternatively the quantum can be measured in utility terms and expressed in an index of health status. However, it is important to note that some methods used to value total costs can also be specified to value intangible costs only. These methods are discussed in section 3.4 on total cost methods. Generally speaking, these methods take the stated preferences of the public in order to make estimates of total or intangible costs (note the methods used to develop health status indices in section can also be based on stated preferences). Specifically, Contingent Valuation is that main method which can be used to measure either total or intangible cost. When this method is used to measure only the intangible costs of injury, it is important to ensure that other sources of the public s perception of total cost is filtered out of their responses. That is, the method has to be tightly specified to gain an accurate representation of the public s perception of the intangible costs Injury Compensation Court awards for injury compensation provide one source of information regarding the collective view of people of the intangible costs of injuries through their deliberations as jurors. These awards can reveal people s collective preferences regarding compensation for intangible costs such as pain and suffering. For example, Rodgers (1998) examined a total of 843 jury awards for pain and suffering in personal product liability cases involving non-fatal injury. The Rodgers study showed four award categories relating to different levels of injury severity as shown in the table below. The average awards for economic loss ranged from $7,000 for the less severe (category 1) injuries to $39,400 for the most severe (category 4) injuries. On average the awards for pain and suffering were orders of magnitude greater than for the associated economic loss, being $35,700 for category 1 and $315,400 for category 4 injuries. 19 Measuring the Total Cost of Injury in New Zealand

21 Awards for Economic Loss, Pain and Suffering by Injury Category Injury Severity Mean Awards ($US) For: % of Total Awards % of Cases % of Awards Economic Loss Pain & Suffering Pain & Suffering Category 1 $7,048 $35,678 84% 16% 8% Category 2 $17,709 $49,889 74% 43% 35% Category 3 $20,747 $76,939 79% 37% 43% Category 4 $39,437 $315,410 89% 3% 14% Average $17,782 $66,158 79% Total 100% 100% Source: Rodgers, 1998 The study also showed that the proportion of the total award that was given for pain and suffering was reasonably constant over the categories, averaging 79% of the total award. Or in other words, the awards for pain and suffering were about four times those of the economic losses. Another observation is that a small number of severe injuries (just 3%) attract a high proportion of the total awards (at 14%) Administrative Compensation Compensation decisions made and schedules set by various public administrations can also be a source of valuable information regarding the revealed preference of collective or administrative decision-makers. This information could be described as derived revealed preferences, because it takes preferences as revealed through a range of administrations determining the value of compensation for one set of pain and suffering, and applies this value to a second set of conditions of pain and suffering. The APC s report on gambling looked at such measures to obtain estimates on pain and suffering from problem gambling 1. The APC s approached was to collect data on the levels of compensation paid across States and Territories of Australia for victims of crime. The APC then surveyed empirical studies of jury awards in the US for pain and suffering. Compensation levels in a number of States are up to $50,000 each for serious harm (Queensland being high at about $75,000). New South Wales offers compensation of $5,000 to $15,000 for cases of chronic psychological or psychiatric disorders that are moderately disabling and $30,000 to $50,000 for disorders that are severely disabling. 1 Appendix J of Australia s Gambling Industries. The Australian Productivity Commission Measuring the Total Cost of Injury in New Zealand

22 The APC then applied these compensation figures to a range of emotional costs that are associated with problem gambling, as shown in the table below. Emotional Costs Associated with Problem Gambling Amounts of $5,000 to $15,000 for: Emotional costs for immediate family of severe problem gamblers Relationship breakdown Depression suffered often to always Amounts of $15,000 to $30,000 for: Divorce or separation Seriously thought of suicide Attempted suicide for immediate family Amounts of $30,000 to $50,000 for: Attempted suicide for the gambler Divorce or separation Attempted suicide for immediate family Source: Australia's Gambling Industries. APC The estimates shown above provide another means of estimating the level and scale of intangible costs as revealed by individuals and/or public administrations. While the jury awards approach taken by Rodgers focused on pain and suffering, the above approach also incorporates social costs such as divorce or separation Health Status Indices Health status indices can be used measure the intangible costs of an injury and despite being a relatively new phenomenon these indices have gained considerable momentum in the field of health economics. This index approach allows different health states to be assessed using a single health status measure such as Quality or Disability Adjusted Life Years (eg QALYs or DALYs). Stated preferences (or utility scores ) are critical variables in calculating these health indices. Utility scores can be elicited by various methods, such as pairwise comparisons, standard gambles, time or person trade-offs, direct ratings from visual analogs and Contingent Valuation. Hence, the impact of an injury event upon health can be mapped across different dimensions such as such as pain and suffering, life expectancy, quality of life and so forth. Thus, the approach enables different dimensions to be measured both separately and in total. 21 Measuring the Total Cost of Injury in New Zealand

23 There are a number of indices available for assigning values to health status. All indices use a scale running from 0 (death) to 1 (no impact). However, each measure is different in both the conceptualisation of health and in the sensitivity to changes in health status. To some extent, these differences reflect that health status indices are a new area of investigation and that there remains some disagreement within the health profession as to the most appropriate measure of health status. However, some differences also reflect that these indices can be designed to focus on certain aspects of health status. Indices as a Measure of Health Status There are a number of indices internationally that can be used assign values to various health states. Eric Nord (1997) identified ten such indices in an OCED review of these measures. the Rosser/Kind Disability/Distress Index the Quality of Well-Being Scale (QWB) the Health Utilities Index, mark II (HU12) the Health Utilities Index, mark III (HU12) the EuroQol instrument (EQ-5D) the Index of Health Related Quality of Life (IHRQOL) the Quality of Life and Health Questionnaire (QLHQ) the Australian Quality of Life Instrument (AQOL) the Years of Healthy Life Measure (YHL) the 15-D Source: A Prescription for Pharmoeconomic Analysis. PHARMAC In New Zealand, the Ministry of Health uses a Disability Adjusted Life Year (DALY) index, which provides a measure of health status derived from adding the years lost to disability and the years lost to premature death. Thus, one DALY represents the loss of one year of healthy life. One of the key variables in the DALY index is the weight attached to disability. These weights were taken from Dutch disability weights estimated by Stouthard et al (1997) using the person trade-off method. These disability weights were chosen in the absence of preferences derived from the population of New Zealand. Some health status indices also apply an age weighting to give more weight to a year of life lived in adulthood than in childhood or old age. The Global Burden of Disease Study completed by Murray and Lopez (1996) is one such example. However, the Ministry of Health does not apply ageweights to the DALY index. 22 Measuring the Total Cost of Injury in New Zealand

24 The health index approach has the obvious advantage of being able to convert intangible costs into a common unit of account. In addition, these intangible costs have the added advantage of having been derived from the revealed preferences of individuals, which is consistent with the notions of utility and consumer welfare in economics. However, the main disadvantage of this approach is that intangible costs are valued using a different unit of account than direct and indirect costs. Therefore, the health index approach does not enable full Cost Benefit Analyses (CBA) of health programs to be undertaken. For this reason the merits of health programs have tended to be assessed using Cost Utility Analysis (CUA). This approach assesses health programs by dividing program expenditures by the QALYs or DALYs saved as a result of the program. Thus, the CUA approach allows for comparisons between programs that impact upon intangible costs, but because direct and indirect costs are not measured, no concept of the net benefit emerges from the assessment. 23 Measuring the Total Cost of Injury in New Zealand